Research on Intelligent Recognition and Analysis Framework of Japanese Katakana in the Background of Big Data and Fuzzy Recognition

With the increase of communication methods, more and more foreign words appear in Japanese communication. Since most of the syllables of these words do not conform to the cognitive schema and usage habits of Japanese katakana, the Japanese vocabulary system is not suitable for some words with more syllables. Vocabulary is abbreviated into katakana abbreviations and romaji abbreviations. The continuous template matching technology is used to analyze the distributed data structure of big data, the information fusion processing of cloud service combined big data is combined with the matching correlation detection method, and the frequent item detection and association rule feature extraction are performed on the high-dimensional fusion data. The method has a prediction accuracy of up to 7.3% and a small error of 6.2%.